A heteroplasmic mitochondrial complex I gene mutation in adult-onset dystonia
Mitochondrial DNA (mtDNA) mutations can cause rare forms of dystonia, but the role of mtDNA mutations in other types of dystonia is not well understood. We now report identification by sequencing, restriction endonuclease analyses, and clonal analyses of a heteroplasmic missense A to G base pair substitution at nucleotide position 3796 (A3796G) in the gene encoding the ND1 subunit of mitochondrial complex I in a patient with adult-onset dystonia, spasticity, and core-type myopathy. The mutation converts a highly conserved threonine to an alanine. The same mutation subsequently was identified in 2 of 74 additional unrelated adult-onset dystonia patients. A muscle biopsy was obtained from 1 of these 2 subjects and this revealed abnormalities of electron transport chain (ETC) activities. The mutation was absent in 64 subjects with early onset dystonia, 82 normal controls, and 65 subjects with Parkinson's disease or multiple system atrophy. The A3796G mutation previously has been reported in 3 of 226 subjects from mitochondrial haplogroup H. Each of the 3 subjects in our study harboring the A3796G mutation was also from haplogroup H. However, a subgroup analysis of haplogroup H subjects from our study indicates that the A3796G mutation is significantly overrepresented among haplogroup H adult-onset dystonia subjects compared with haplogroup H controls (P<0.01). This difference remains significant even after excluding the index patient (P=0.04). These data suggest that, among haplogroup H subjects, the presence of the A3796G mutation increases the risk of developing adult-onset dystonia.
KeywordsGenetics Mitochondria Haplogroup ND1
Genetic causes of some forms of early onset dystonia have been identified [1, 2, 3], but the cause of the most-common form of dystonia, adult-onset sporadic dystonia, remains unknown in the majority of cases. Several lines of evidence implicate a role for mitochondrial dysfunction in the pathogenesis of dystonia. 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP), an inhibitor of mitochondrial complex I, induces transient dystonia in baboons . A mitochondrial toxin that inhibits complex II, 3-nitropropionic acid, induces a delayed dystonia in monkeys , and has induced dystonia following accidental poisoning in humans [6, 7]. In human subjects, platelet mitochondrial complex I activity appears to be decreased in idiopathic dystonia in some studies [8, 9, 10].
Although the origin of complex I dysfunction in dystonia is not known in most cases, genetic mutations affecting mitochondrial function have been associated with rare forms of dystonia. Mutations in a nuclear gene encoding a mitochondrial import protein (deafness dystonia peptide) have been identified as the cause of X-linked dystonia-deafness syndrome [11, 12]. Female carriers of a mutation in this gene can present with adult-onset focal dystonia . Mitochondrial DNA (mtDNA) mutations also have been linked to dystonia. A mutation at position 14459 in the ND6 subunit of mitochondrial complex I was identified in several families with dystonia and Leber's hereditary optic neuropathy (LHON) [14, 15]. Dystonia has also been reported in association with other LHON-associated mtDNA mutations [16, 17]. Dystonia can be a presenting feature of the 3243 MELAS (mitochondrial encephalomyopathy, lactic acidosis and stroke-like episodes) mutation . Dystonia is also a common feature of Leigh's syndrome, which is often associated with mtDNA mutations [19, 20, 21], and has been reported in association with Kearns-Sayre syndrome . A mutation altering the initiating methionine of the ND1 subunit of complex I was reported in a patient with focal dystonia and maternally inherited cataracts .
These examples demonstrate that dystonia can be a major clinical manifestation of a mtDNA mutation. The current study expands the spectrum of mtDNA mutations associated with dystonia, and suggests that mtDNA mutations may play a role in a significant subset of adult-onset dystonia patients.
Materials and methods
Sample information: dystonia subjects
Age of onset
No or ?
DNA was isolated by standard proteinase K and sodium dodecyl sulfate digestion followed by phenol and chloroform extractions. Polymerase chain reaction (PCR) amplification of mtDNA and sequencing on an ABI 377 or 3700 automated sequencer (Applied Biosystems) were performed as previously described . PCR for restriction digests was performed with primers at nucleotide positions (5′ to 3′) 3601–3616 and 3953–3939. Restriction digests were performed with AciI (New England Biolabs, Beverly, Mass., USA) and analyzed by ultraviolet illumination of a 2% agarose gel permeated with ethidium bromide. The A3796G mutation creates a single restriction site for this enzyme, causing the normal 352-base pair PCR product to be cut into 193- and 159-base pair fragments. Other PCR and sequencing primers have been published previously . The chi-squared test was used for analyzing differences in mutation frequencies between groups. Cloning was performed using the TOPO TA cloning kit (Invitrogen, Carlsbad, Calif., USA) according to the manufacturer's directions. PCR prior to cloning was performed with the HF2 high-fidelity PCR kit (Clontech, Palo Alto, Calif., USA). Other PCR reactions were performed with Amplitaq Gold (Roche, Branchburg, N.J., USA). M13 forward (5′CAGGAAACAGCTATGAC3′) and reverse (5′GTAAAACGACGGCCAG3′) primers, corresponding to the vector flanking the region of the PCR product insertion site, were used in post-cloning PCR. For analysis of a polymorphic site at 7028, PCR amplification was performed with primers at nucleotide positions (5′ to 3′) 6924–6938 and 7470–7453. A "T" is present at this site in most haplogroups, creating an additional AluI (New England Biolabs) restriction site, resulting in bands of 397, 103, and 29 base pairs. Approximately 95% of haplogroup H subjects carry a "C" at this site , which eliminates this AluI restriction site, yielding bands of 397 and 132 base pairs.
Analysis of electron transport chain activity
Electron transport chain (ETC) activities were measured on a skeletal muscle biopsy from a dystonia patient harboring a homoplasmic A3796G mutation, in a commercial laboratory as previously described .
We investigated the role of mtDNA mutations in adult-onset dystonia initially by sequencing the mitochondrial complex I genes in a patient with focal dystonia which began at 34 years of age (dystonia patient 1). She initially developed oral-mandibular dystonia that later also involved the neck muscles bilaterally. In her 40s, she developed weakness, spasticity, hyperreflexia, and extensor plantar responses. She died at 47 years of age of aspiration pneumonia and congestive heart failure secondary to aortic stenosis. Family history was notable for Alzheimer's disease in her mother in her 70s, but was negative for dystonia. Two female siblings were normal by history and neurological examination. One male sibling, as well as the patient's son and daughter, were normal by history.
Two skeletal muscle biopsies were performed, one ante mortem and one at the time of autopsy. The biopsies showed myopathic changes with variability of fiber size, with small atrophic fibers and increased internalized nuclei, nuclear clumping, and fibers with sarcomeric disorganization of a core and mini-core type. Changes typical of a mitochondrial cytopathy were lacking, with reactive cytochrome c oxidase and succinate dehydrogenase staining. However, cytochrome c oxidase and NADH histochemical reactions revealed discrete patchy areas of decreased staining within otherwise positively staining individual type 1 fibers. The modified Gomori trichrome stain showed no ragged red fibers. Electron-microscopic examination did not reveal paracrystalline inclusions or structural abnormalities of mitochondria. Areas of sarcoplasmic disorganization with Z-band streaming consistent with cores and core-like areas were seen. Cytoplasmic bodies were also seen on the autopsy muscle specimen. Magnetic resonance imaging of the brain revealed only generalized atrophy and mild nonspecific white matter changes. Postmortem examination of the brain showed an acute hemorrhagic infarction in the left parietal cortex, as well as widespread acute (agonal) hypoxic-ischemic neuronal injury. No abnormalities were noted in the basal ganglia or the spinal cord.
Evolutionary conservation of the amino acid altered by the A3796G mutation
DNA samples from 159 additional unrelated dystonia subjects were screened for the A3796G mutation by the AciI restriction endonuclease analysis (Table 2). Two of these additional dystonia patients tested positive for the A3796G mutation by restriction digestion (Fig. 1). Patient 2 was a woman who had been diagnosed with secondary dystonia based on progressive right leg posturing and dystonic tremor beginning at 29 years of age, 4 months after a fall with injury to the back. The dystonia later spread to her right hand, arm, and face, and then to the left foot. Magnetic resonance imaging of the brain and spine revealed no abnormalities. Family history was negative for dystonia, although notable for stroke in her mother in her late 20s. Patient 3 was a woman with chronic anxiety who developed tardive dystonia beginning at 42 years of age after chronic neuroleptic exposure. The dystonia persisted despite discontinuation of the neuroleptic. She initially presented with lower face dystonia, which later became multi-focal, involving the face, pharynx, neck, and both arms. Speech and swallowing were impaired. No other neurological abnormalities were noted on examination and family history was negative for dystonia.
Each of the 3 patients testing positive for the A3796G mutation belonged to mitochondrial haplogroup H based on the absence of the G11719A polymorphism and the presence of the T7028C polymorphism . The A3796G mutation was recently reported in 3 of 226 haplogroup H subjects , representing the first report in the literature of this mutation. To determine if the subgroup of haplogroup H subjects with the 3796 mutation is at increased risk for developing dystonia compared with haplogroup H subjects lacking this mutation, we performed a subgroup analysis of our haplogroup H dystonia patients and controls, defined here by the presence of a "C" at 7028 detected by restriction endonuclease analysis. The A3796G mutation was present in a greater percentage of the haplogroup H dystonia patients (present in 3 of 42 cases) compared with haplogroup H controls (present in none of 50 controls, P=0.05). This difference is not statistically significant if the index case is excluded (P=0.11). We next performed a meta-analysis, including our data as well as the 226 haplogroup H subjects screened by Herrnstadt et al. . In this case, the difference was clearly significant (P<0.01) if all 3 of the 3,796 positive dystonia subjects were included, and approached significance when the index case was excluded (P=0.06). Exclusion of the index case accounts for the fact that the mutation was identified in the first patient by a general screen conducted by sequencing all of the mtDNA-encoded complex I genes, whereas the remaining 159 patients were analyzed by a restriction endonuclease assay for the A3796G mutation with the a priori hypothesis that this specific mutation would be found at a higher frequency in the dystonia subjects. If the analyses are restricted to haplogroup H adult-onset dystonia patients (onset at age 21 years or older), then the mutation was present in 2 of 24 cases (excluding the index case), representing a significant overrepresentation of 3,796 positive subjects among the haplogroup H dystonia subjects compared with the 50 haplogroup H controls from our study (P=0.04) or compared with the combined group of haplogroup H controls in our study plus the 226 haplogroup H subjects reported by Herrnstadt et al.  (P<0.01).
A skeletal muscle biopsy (left quadriceps) was performed on patient 2. This revealed increased variation in fiber size with scattered small fibers. Rare basophilic fibers were present. NADH reductase, COX, succinate dehydrogenase, and modified Gomori's trichrome (oil red O, periodic acid-Schiff and ATPase) stains did not reveal changes suggestive of a mitochondrial cytopathy. Electron microscopy showed mitochondria that were normal in number, distribution, and morphology.
Electron transport chain (ETC) activities
NADH-cytochrome c reductase
I and III
Succinate-cytochrome c reductase
II and III
Decylubiquinol-cytochrome c reductase
Cytochrome c oxidase
We have identified a heteroplasmic complex I gene missense mutation in 3 patients with adult-onset dystonia. In the first patient, although the combination of spasticity, dystonia, and myopathy initially raised suspicion of a mitochondrial disorder, pathology revealed a core-type myopathy without specific changes typical of some mitochondrial genetic disorders. Unfortunately, tissue suitable for ETC activity measurements was unavailable. However, such in vitro assays do not consistently show defects, even in the setting of well-accepted pathogenic mutations. Similarly, although the mutation appeared to be incompletely penetrant, even in the families of the 2 patients harboring a homoplasmic mutation, incomplete penetrance is a common feature of some known pathogenic mtDNA mutations, such as the G11778A complex I mutation associated with LHON .
Several factors indicate that the mutation may be pathogenic, although it remains important to confirm the association in future studies. The mutation was heteroplasmic in the first family identified to harbor the mutation. Heteroplasmy is thought to represent a transitional state commonly seen in pathogenic mtDNA mutations [28, 29]. Thus, the presence of heteroplasmy for the A3796G mutation in 1 of the dystonia subjects would be surprising for a non-pathogenic polymorphism, and supports the hypothesis that this mutation is pathogenic. Although the mutation was present in asymptomatic maternal relatives of patient 1, incomplete penetrance with sporadic expression is common among known pathogenic mtDNA mutations . Furthermore, the mutation was absent in 82 normal controls and in 65 subjects with Parkinson's disease or multiple system atrophy, and alters an amino acid that is highly conserved evolutionarily. Finally, biochemical studies support an abnormality of mitochondrial energy metabolism (Table 3). ETC analyses in skeletal muscle from patient 2 are suggestive of impaired activity in the distal parts of complex I and possibly a mild defect in complex IV. Elevated long-chain acylcarnitines, although not diagnostic, also suggest an abnormality of mitochondrial metabolism (Charles L. Hoppel, personal communication). A potential concern in any genetic association study, particularly one involving mtDNA (which is highly polymorphic), is a spurious association due to imperfect matching of the disease and control groups. The A3796G mutation recently was reported in 4 of more than 500 subjects  by complete sequencing of the mitochondrial genomes, including 3 of 226 subjects from mitochondrial haplogroup H, representing 1.3% of this subgroup. The 3 patients reported here who tested positive for the A3796G mutation also belong to haplogroup H. This raises the possibility that the apparent association of this mutation with dystonia in our subjects could be due to an increased frequency of haplogroup H among the dystonia subjects compared with the controls. This possibility was addressed in two ways. First, the frequency of the haplogroup H-associated G7028C polymorphism was analyzed in the control and dystonia subjects. This polymorphism was present in 31% of the dystonia subjects and in 39% of the controls screened for this polymorphism, indicating that a relative under-representation of haplogroup H did not occur in our control group. Second, we performed a subgroup analysis of the haplogroup H dystonia patients and controls. In this subgroup, the A3796G mutation was significantly more frequent among adult-onset dystonia subjects compared with controls (P<0.01), even after exclusion of the index case (P=0.04). The significant association of the A3796G mutation with adult-onset dystonia, even when patients are compared with haplogroup-matched controls, suggests that the A3796G mutation increases the risk of adult-onset dystonia among haplogroup H subjects. It remains possible that the A3796G mutation could be a marker for another haplogroup H-associated mtDNA mutation that is pathogenic with respect to dystonia risk. However, in patient 1, who harbored a heteroplasmic mutation, no other mutations were identified by sequencing each of the seven of the mitochondrial-encoded complex I genes.
The distinction between neutral "polymorphisms" and pathogenic "mutations" has become increasingly blurred, particularly among mtDNA sequence variants. For example, haplogroup J-associated polymorphisms at 4216 and 13708 are not pathogenic on their own, but increase the penetrance of primary LHON mutations [31, 32, 33]. Haplogroup J-associated polymorphisms also appear to influence the risk of developing multiple sclerosis or Devic's disease [34, 35]. Similarly, our data suggest that the haplogroup H-associated A3796G mutation may influence the risk of developing adult-onset dystonia, presumably through interactions with other genetic and environmental factors.
The alteration by the A3796G mutation of a conserved amino acid in a mitochondrial complex I gene is consistent with reports of complex I dysfunction in patients with focal dystonia [8, 9]. One study failed to demonstrate a complex I defect in human lung carcinoma-derived cybrid cell lines expressing mtDNA from nine patients with focal dystonia, and has been interpreted as evidence against a role for mtDNA mutations in dystonia . However, because the biochemical expression of mtDNA mutations is highly dependent on both nuclear background and tissue type [37, 38], the lack of detection of a complex I defect in this single non-neuronal cell type is inconclusive as to the origin of the complex I defect in dystonia patients.
Mutations in several distinct nuclear and mitochondrial genes have been identified in association with rare forms of familial dystonia, and others have been localized [1, 2, 3]. The A3796G mutation reported here is unique as it was identified in 3 unrelated adult-onset dystonia patients who lacked a family history of dystonia. The frequency of the mutation seen in this study cannot be extrapolated to the general population of dystonia patients, as this group included relatively high percentages of early onset, familial, and secondary dystonias. Although the dystonia was focal in onset in all 3 patients positive for the A3796G mutation, in each patient it subsequently spread to involve other areas. It remains to be seen if this mutation plays a role in typical persistently focal adult-onset dystonia. The significance of the identification of the mutation in a patient with tardive dystonia and another with post-traumatic dystonia is also uncertain. This could represent coincidence. Alternatively, the mutation may predispose to the development of dystonia in conjunction with certain environmental exposures, such as trauma or neuroleptic exposure. Interactions between mtDNA mutations and environmental factors are well established in some cases, such as the 1555 ribosomal RNA mutation and aminoglycoside-induced deafness [39, 40]. Genetic-environmental interactions could at least partially account for the incomplete penetrance seen with the A3796G mutation and with other mtDNA mutations. Screening of additional patients with primary and secondary dystonias will be necessary to clarify this issue for the A3796G mutation.
These data suggest that, among haplogroup H subjects, the presence of the A3796G mitochondrial complex I gene mutation, together with other genetic and environmental factors, increases the risk of developing adult-onset dystonia. This has significant implications, and thus it will be important to confirm the association in additional populations of dystonia patients. The association of the A3796G mutation with dystonia further implicates mitochondrial complex I dysfunction in the pathogenesis of dystonia, and raises the possibility that this genetically defined subgroup of patients may benefit from strategies to improve mitochondrial function or reduce oxidative stress.
This study was supported by the Ann B. and Norman A. Bikalis Fund for Dystonia Research (D.K.S.), the Bachmann-Strauss Dystonia Foundation (M.F.B.), and by the Medical Research Service, Department of Veteran Affairs (M.E.C.). We thank Dr. C.L. Hoppel for assistance with electron transport chain analyses.
- 2.Jarman PR, Grosso N del, Valente EM, Leube B, Cassetta E, Bentivoglio AR, Waddy HM, Uitti RJ, Maraganore DM, Albanese A, Frontali M, Auburger G, Bressman SB, Wood NW, Nygaard TG. (1999) Primary torsion dystonia: the search for genes is not over. J Neurol Neurosurg Psychiatry 67:395–397PubMedGoogle Scholar
- 11.Jin H, May M, Tranebjaerg L, Kendall E, Fontan G, Jackson J, Subramony SH, Arena F, Lubs H, Smith S, Stevenson R, Schwartz C, Vetrie D (1996) A novel X-linked gene, DDP, shows mutations in families with deafness (DFN-1), dystonia, mental deficiency and blindness. Nat Genet 14:177–180PubMedGoogle Scholar
- 23.Simon DK, Tarnopolsky MA, Greenamyre JT, Johns DR (2001) A frameshift mitochondrial complex I gene mutation in a patient with dystonia and cataracts: is the mutation pathogenic? J Med Genet 38:58–61Google Scholar
- 24.Herrnstadt C, Elson JL, Fahy E, Preston G, Turnbull DM, Anderson C, Ghosh SS, Olefsky JM, Beal MF, Davis RE, Howell N (2002) Reduced-median network analysis of complete mitochondrial DNA coding-region sequences for the major African, Asian, and European haplogroups. Am J Hum Genet 70:1152–1171Google Scholar
- 25.Hoppel CL, Kerr DS, Dahms B, Roessmann U (1987) Deficiency of the reduced nicotinamide adenine dinucleotide dehydrogenase component of complex I of mitochondrial electron transport. Fatal infantile lactic acidosis and hypermetabolism with skeletal-cardiac myopathy and encephalopathy. J Clin Invest 80:71–77PubMedGoogle Scholar
- 33.Torroni A, Petrozzi M, D'Urbano L, Sellitto D, Zeviani M, Carrara F, Carducci C, Leuzzi V, Carelli V, Barboni P, De Negri A, Scozzari R (1997) Haplotype and phylogenetic analyses suggest that one European-specific mtDNA background plays a role in the expression of Leber hereditary optic neuropathy by increasing the penetrance of the primary mutations 11778 and 14484. Am J Hum Genet 60:1107–1121PubMedGoogle Scholar
- 41.MITOMAP (2001) A Human Mitochondrial Genome Database. Center for Molecular Medicine, Emory University, Atlanta, Ga., USAhttp://www.gen.emory.edu/mitomap.html